Hyper-, hypokalemia with SGLT-2 inhibitors related to eGFR decrease after initiation

Patients with type 2 diabetes who have large decreases in estimated glomerular filtration rate (eGFR) after starting sodium-glucose cotransporter-2 (SGLT-2) inhibitor therapy may be at higher risk for hyper- and hypokalemia, according to a recent study.

Researchers in Taiwan performed a retrospective observational study of patients with type 2 diabetes who had baseline and follow-up eGFR data available after four to 12 weeks of treatment with an SGLT-2 inhibitor. Serum potassium homeostasis depends largely on kidney excretion, and it is unknown whether an abrupt decrease in eGFR after starting SGLT-2 inhibitor therapy is related to dyskalemia risk, the researchers noted.

The goal of the study was to determine the effect of different levels of change in eGFR on serum potassium homeostasis after SGLT-2 inhibitors were started. The primary outcomes were hyperkalemia events, defined as a serum potassium measurement of 5.5 mmol/L or greater or a prescription of potassium binder, and hypokalemia events, defined as a serum potassium measurement less than 3.0 mmol/L or a prescription of potassium supplement during the follow-up period. Results were published April 30 by the Journal of the American Heart Association.

Patients with type 2 diabetes were included in the trial if they received a first prescription of an SGLT-2 inhibitor on or after June 1, 2016, and had potassium measurements from before and after an SGLT-2 prescription. The study period ended on Dec. 31, 2018. Among all study participants, treatment with SGLT-2 inhibitors was associated with an initial mean eGFR decline of −3.5 mL/min per 1.73 m². In 36.7% of patients, there was no eGFR decline, while 57.9% had a decline of 0% to 30% and 5.4% had a decline of more than 30%.

Patients with an initial eGFR decline of more than 30% had more variability in serum potassium levels versus those with no initial eGFR decline. A decline of more than 30% in eGFR after SGLT-2 inhibitor treatment was associated with higher risk for hyperkalemia (adjusted hazard ratio [HR], 4.59; 95% CI, 2.28 to 9.26) or use of a potassium binder (adjusted HR, 2.65; 95% CI, 1.78 to 3.95) and hypokalemia (adjusted HR, 3.21; 95% CI, 1.90 to 5.42) or use of potassium supplements (adjusted HR, 1.87; 95% CI, 1.37 to 2.56) after multivariate adjustment.

The researchers noted that the study could have been affected by residual or unmeasurable confounding, among other limitations. They concluded that in patients with type 2 diabetes, treatment with SGLT-2 inhibitors decreased eGFR by a small amount overall.

"There was a higher incidence of hyperkalemic and hypokalemic events in a small number of patients, especially those whose eGFR dropped significantly at the start of treatment," the authors wrote. "Conversely, a modest early drop in eGFR (eg <30%) following [SGLT-2 inhibitor] treatment was not associated with an increase in the risk of hypokalemia or hyperkalemia." Physicians should be aware that the eGFR trough and resulting changes in serum potassium levels appear to occur shortly after SGLT-2 inhibitors are initiated, they said.